4c72a20ba2
[*] Win32: required to prevent leaks when invoked by Qt (todo: proper journaling of pending actions)
2023-06-24 17:41:09 +01:00
74e036b809
[*] NT: Switch the event primitives used in old AuLoop (Aurora::Loop - no IO) subsystem
2023-06-23 07:56:40 +01:00
573031cd96
[*] idiot-proofing memory ownership
2023-06-11 15:49:39 +01:00
1f505fcf02
[+] Added explicit stream interface cache reset path under proto stacks
...
(amended: missed one discard)
2023-06-11 15:30:14 +01:00
4a584ce73a
[+] Future proof IProtocolInterceptorEx with an optional pParent in the only Aurora Interface methods last argument
2023-06-10 20:33:19 +01:00
301f1a6025
[+] yes
2023-06-04 22:19:57 +01:00
8b36562cf0
[+] why not?
2023-06-04 21:18:44 +01:00
78091bf84b
[+] also gonna impl this flush
2023-06-04 20:52:22 +01:00
bbd4679bed
[+] AuIO::IOPipeCallback::bCloseWriter
2023-06-04 20:49:13 +01:00
bc203d8cbc
[+] ProtocolPiece::GetParent
2023-06-04 19:07:26 +01:00
c9ed4b09b9
[+] Implement IStream[Read/Write].close for various IO subsystem interfaces
2023-06-04 19:02:33 +01:00
81b3550825
[+] AURT_ADD_USR_DATA usage across some relevant contexts
2023-06-04 17:28:29 +01:00
b7de15d78e
[*] tweak LoopQueue.nt.cpp
...
[*] bug fixes for running Qt alongside GTK and other loops
2023-05-16 17:11:59 +01:00
e2f16091d9
[*] amend write ahead check
2023-05-11 18:31:47 +01:00
ac4968512b
[*] cleaned up an an unwanted network error from the logs by not flushing hello packets should the socket be in an unconnected state
2023-05-11 17:46:49 +01:00
5ded6a751f
[+] STATUS_CONNECTION_REFUSED for some lower-level firewall drivers under Windows
2023-05-11 17:38:14 +01:00
1fa433540b
[*] Experimental lower-latency IO pipes
...
[*] revert win32 net tx hardening - do not use the Reset function to null pSocket
[*] fix a bytebuffer circular buffer path
[*] update how TLS protocol stacks can snap the stack
2023-05-11 16:05:00 +01:00
2fdf730263
[*] Stats now return nanoseconds timestamps
2023-05-03 08:57:21 +01:00
6683d3cfdc
[+] IProtocolPiece.GetNextPiece
...
[+] IProtocolPiece.ReallocateDrainBuffer
[*] Harden protocol piece use after destroy
2023-04-29 07:00:29 +01:00
da69428d7e
[+] IProtocolStack::PrependInterceptorEx
...
[+] IProtocolStack::PrependSingleFrameProcessor
[+] IProtocolStack::PrependSingleFrameProcessorEx
[+] IProtocolStack::GetArrayOfInterceptors
[+] IProtocolStack::GetInterceptorAtIndex
[+] IProtocolPiece::ToInputWriter
[+] IProtocolPiece::GetExtendedInterceptor
[+] IProtocolPiece::GetShortPipeInterceptor
2023-04-28 02:41:45 +01:00
f8219542c5
[*] precache socket network state errors here before the last transaction is released
2023-04-27 10:44:15 +01:00
eb4ee3143d
[*] bug: missing link
2023-04-27 10:26:41 +01:00
f1e9a7262b
[*] pipe processors with a stack attached to them should destroy the stack to forcefully break circular references in user code
2023-04-27 06:10:09 +01:00
f70d6392d9
[*] harden
2023-04-27 01:34:29 +01:00
efadd964be
[*] Improved disconnect reason reporting: HasErrorCode under the NT network async stream object
2023-04-26 20:32:25 +01:00
4b79aa902b
[*] i dont think this path should ever be hit, but just to be certain, lets check the networks output channel translation object for an error if the socketbase doesnt know of the last fatal error code
2023-04-26 20:22:11 +01:00
b8540c77a7
[*] improved network error reporting
2023-04-26 20:13:17 +01:00
3f0368b99d
[*] cache these adapter allocations away
2023-04-26 07:37:14 +01:00
30b0c2e0ba
[*] Preemptive GetOverlappedResult fast-paths under WaitMultiple2 that rely on the alertable yield under next read/write cannot be trusted if the read count is zero.
...
Direct IO-specific quirk?
Windows 7 specific quirk?
Direct IO under Win7 specific quirk?
idk, but this fixes it
2023-04-26 07:25:45 +01:00
22c4d25759
[*] amend cdf5a499, bKillPipeOnFirstRootLevelFalse not respected
2023-04-25 12:57:15 +01:00
cdf5a499cb
[*] Improved pipe work awareness under protocol stack to allow for propagation of error conditions such that the top most interceptor can kill the pipe work
2023-04-25 12:13:16 +01:00
e0d7bbea01
[*] stream regression
2023-04-25 10:09:37 +01:00
664eb270dc
[+] IIOPipeWork::GetBytesProcessedInterframe
...
[*] IOPipeProcessors will now cap reported progression to the requested length if the requested page size would mean we overrun it.
2023-04-25 07:38:36 +01:00
2a33d61e63
[*] further deprecate high res clock
...
[*] further posix resolution reporting
2023-04-22 22:58:20 +01:00
36c163d90c
[*] 2af2eadf cont
2023-04-19 23:57:56 +01:00
d7d514c6b7
[*] protocol-stack ticks now check against the buffers error flags to ensure Writes/Reads weren't being silently dropped in user code
2023-04-19 23:46:56 +01:00
2af2eadfa8
[*] Resolve: protocol-stack link bug (multiple append failure / missing next link)
2023-04-19 23:35:48 +01:00
00129255f8
[*] changed this to be a soft-failure
2023-04-19 04:03:49 +01:00
53a29a6c13
[*] flag consistency
2023-04-16 23:41:06 +01:00
4cac821fbb
[+] Added critical log level
...
[*] Fix win32 DELETE bug
2023-04-16 23:38:31 +01:00
a550a6f58f
[*] Harden normalize path
2023-03-05 09:16:09 +00:00
1e4971342d
[*] big dumb
2023-02-04 19:46:35 +00:00
f43251c8fc
[+] AuNet::ISocketChannelEventListener
...
[+] AuFS::UpdateTimes
[+] AuFS::UpdateFileTimes
[+] AuFS::CompressEx
[*] AuFS::Compress now rejects files that look to be already compressed
[+] AuFS::DecompressEx
[+] AuFS::Create
[+] AuFS::WriteNewFile
[+] AuFS::WriteNewString
[+] AuFs::FileAttrsList
[+] AuFs::FileAttrsGet
[+] AuFs::FileAttrsSet
[+] DirectoryLogger::uMaxLogsOrZeroBeforeCompress
[+] ISocketChannel.AddEventListener
[+] ISocketChannel.AddEventListener
[+] DirectoryLogger.uMaxLogsOrZeroBeforeCompress
[*] Fix UNIX regression
[*] Fix up stream socket channel realloc IPC
[*] Fix shutdown regression in pretty much everything thanks to 8ff81df1's dumbass fix
(fixes fence regression on shutdown)
[*] Fix DirDeleterEx formatting of reported failed paths
[*] Fix up file not truncated if already exists bugs. Extended and alternative apis added.
[*] Fix ICompressionStream::ReadEx returning the wrong read value
[+] Legacy compression API can now self-correct once newer stream processor objects are added
2023-02-04 19:43:01 +00:00
2474566d80
[*] two hotfixes
2023-01-29 23:03:54 +00:00
04aca5fcf2
[+] Aurora::IO::Net::NetSocketConnectByHost
...
[+] Aurora::IO::FS::DirDeleterEx
[+] Aurora::IO::Compress
[+] Aurora::IO::Decompress
[*] Aurora::Memory::ByteBuffer zero-alloc fixes
[*] Aurora::Memory::ByteBuffer linear read of begin/end should return (`const AuUInt8 *`)'s
[*] Changed NT file CREATE flags
[*] Fix linux regression
[*] Update logger sink DirLogArchive
... [+] DirectoryLogger::uMaxLogsOrZeroBeforeDelete
... [+] DirectoryLogger::uMaxCumulativeFileSizeInMiBOrZeroBeforeDelete
... [+] DirectoryLogger::uMaxCumulativeFileSizeInMiBOrZeroBeforeCompress
... [+] DirectoryLogger::uMaxFileTimeInDeltaMSOrZeroBeforeCompress
... [+] DirectoryLogger::uMaxFileTimeInDeltaMSOrZeroBeforeDelete
[*] FIX: BufferedLineReader was taking the wrong end head
(prep) LZMACompressor
[*] Updated build-script for LZMA (when i can be bothered to impl it)
(prep) FSOverlappedUtilities
(prep) FSDefaultOverlappedWorkerThread | default worker pool / apc dispatcher / auasync dispatcher concept for higher level overlapped ops
(stub) [+] Aurora::IO::FS::OverlappedForceDelegatedIO
(stub) [+] Aurora::IO::FS::OverlappedCompress
(stub) [+] Aurora::IO::FS::OverlappedDecompress
(stub) [+] Aurora::IO::FS::OverlappedWrite
(stub) [+] Aurora::IO::FS::OverlappedRead
(stub) [+] Aurora::IO::FS::OverlappedStat
(stub) [+] Aurora::IO::FS::OverlappedCopy
(stub) [+] Aurora::IO::FS::OverlappedRelink
(stub) [+] Aurora::IO::FS::OverlappedTrustFile
(stub) [+] Aurora::IO::FS::OverlappedBlockFile
(stub) [+] Aurora::IO::FS::OverlappedUnblockFile
(stub) [+] Aurora::IO::FS::OverlappedDelete
2023-01-26 21:43:19 +00:00
bf8c1eb8c7
[+] AuFS::TrustFile
...
[+] AuFS::UnblockFile
[+] AuFS::TrustFile
[*] Fixup UNIX open logic a bit
(amended. i did not mess up octals >:( )
2023-01-25 16:50:53 +00:00
d4dfe22c6c
[+] Always expand directory tree when attempting to copy or move files, just like every other FIO create operation.
...
[*] ~~Fix improper yield BlockUntilComplete on NewFence~~, kind of, not really. Wasn't really a bug. Might've introduced more overhead. Might revert.
2023-01-23 21:18:58 +00:00
0cdbc34c06
[*] Optimize allocations out of RWLock
...
[*] Fix linux regression
2022-12-29 09:42:02 +00:00
2209aeb7a8
[+] Linux: semaphores and mutexes directly over futexes. Move UNIX pthread condvar mutex into the condvar mutex class.
...
[*] BSD: Rewrote fundamentally flawed pthread_mutex class code to use MONOTONIC clock time
[+] Linus SwInfo: Added enterprise check for RedHat
2022-12-28 23:44:45 +00:00
a1e77a471b
[*] Dumb path bug
2022-12-21 14:27:26 +00:00
